{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:D6UF3BN6GP3ZOBNKYZEMNUOLFJ","short_pith_number":"pith:D6UF3BN6","schema_version":"1.0","canonical_sha256":"1fa85d85be33f79705aac648c6d1cb2a5e2c61987ad8ce54ec8fab90555bd4b9","source":{"kind":"arxiv","id":"2509.14760","version":3},"attestation_state":"computed","paper":{"title":"Reasoning over Boundaries: Enhancing Specification Alignment via Test-time Deliberation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bo Li, Dongrui Liu, Haoran Zhang, Xuyang Hu, Yafu Li, Yu Cheng, Zhilin Wang","submitted_at":"2025-09-18T09:08:53Z","abstract_excerpt":"Large language models (LLMs) are increasingly applied in diverse real-world scenarios, each governed by bespoke behavioral and safety specifications (spec) custom-tailored by users or organizations. These spec, categorized into safety-spec and behavioral-spec, vary across scenarios and evolve with changing preferences and requirements. We formalize this challenge as specification alignment, focusing on LLMs' ability to follow dynamic, scenario-specific spec from both behavioral and safety perspectives. To address this challenge, we propose Align3, a lightweight method that employs Test-Time De"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2509.14760","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-09-18T09:08:53Z","cross_cats_sorted":[],"title_canon_sha256":"20ee7bd98ecfa6840f8d63543704eac117c2403b6a17122b1d5622f69b9aac72","abstract_canon_sha256":"e4439f3438c55c726d9f141db1bf5b000d763d4a7b9b5da5704136927c83651c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-04T01:08:32.074188Z","signature_b64":"Z7msNkRLyyjlfVt8XA5EmfQjFZJuPpl3OwB+g+o4CY1RVMkhCjgjHnIn17bnLstbZTBXMGAH9hHlay8/Eu7wCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1fa85d85be33f79705aac648c6d1cb2a5e2c61987ad8ce54ec8fab90555bd4b9","last_reissued_at":"2026-06-04T01:08:32.073488Z","signature_status":"signed_v1","first_computed_at":"2026-06-04T01:08:32.073488Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reasoning over Boundaries: Enhancing Specification Alignment via Test-time Deliberation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bo Li, Dongrui Liu, Haoran Zhang, Xuyang Hu, Yafu Li, Yu Cheng, Zhilin Wang","submitted_at":"2025-09-18T09:08:53Z","abstract_excerpt":"Large language models (LLMs) are increasingly applied in diverse real-world scenarios, each governed by bespoke behavioral and safety specifications (spec) custom-tailored by users or organizations. These spec, categorized into safety-spec and behavioral-spec, vary across scenarios and evolve with changing preferences and requirements. We formalize this challenge as specification alignment, focusing on LLMs' ability to follow dynamic, scenario-specific spec from both behavioral and safety perspectives. To address this challenge, we propose Align3, a lightweight method that employs Test-Time De"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.14760","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2509.14760/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2509.14760","created_at":"2026-06-04T01:08:32.073587+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.14760v3","created_at":"2026-06-04T01:08:32.073587+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.14760","created_at":"2026-06-04T01:08:32.073587+00:00"},{"alias_kind":"pith_short_12","alias_value":"D6UF3BN6GP3Z","created_at":"2026-06-04T01:08:32.073587+00:00"},{"alias_kind":"pith_short_16","alias_value":"D6UF3BN6GP3ZOBNK","created_at":"2026-06-04T01:08:32.073587+00:00"},{"alias_kind":"pith_short_8","alias_value":"D6UF3BN6","created_at":"2026-06-04T01:08:32.073587+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/D6UF3BN6GP3ZOBNKYZEMNUOLFJ","json":"https://pith.science/pith/D6UF3BN6GP3ZOBNKYZEMNUOLFJ.json","graph_json":"https://pith.science/api/pith-number/D6UF3BN6GP3ZOBNKYZEMNUOLFJ/graph.json","events_json":"https://pith.science/api/pith-number/D6UF3BN6GP3ZOBNKYZEMNUOLFJ/events.json","paper":"https://pith.science/paper/D6UF3BN6"},"agent_actions":{"view_html":"https://pith.science/pith/D6UF3BN6GP3ZOBNKYZEMNUOLFJ","download_json":"https://pith.science/pith/D6UF3BN6GP3ZOBNKYZEMNUOLFJ.json","view_paper":"https://pith.science/paper/D6UF3BN6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.14760&json=true","fetch_graph":"https://pith.science/api/pith-number/D6UF3BN6GP3ZOBNKYZEMNUOLFJ/graph.json","fetch_events":"https://pith.science/api/pith-number/D6UF3BN6GP3ZOBNKYZEMNUOLFJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/D6UF3BN6GP3ZOBNKYZEMNUOLFJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/D6UF3BN6GP3ZOBNKYZEMNUOLFJ/action/storage_attestation","attest_author":"https://pith.science/pith/D6UF3BN6GP3ZOBNKYZEMNUOLFJ/action/author_attestation","sign_citation":"https://pith.science/pith/D6UF3BN6GP3ZOBNKYZEMNUOLFJ/action/citation_signature","submit_replication":"https://pith.science/pith/D6UF3BN6GP3ZOBNKYZEMNUOLFJ/action/replication_record"}},"created_at":"2026-06-04T01:08:32.073587+00:00","updated_at":"2026-06-04T01:08:32.073587+00:00"}